Credit Card Fraud Detection Using Machine Learning: An AI-Driven Approach for Financial Security
AI-Driven Fraud Detection Presentation — IEEE 2026Abstract
Credit card fraud poses a major challenge to financial institutions, causing severe economic losses and eroding consumer trust.
This study applies supervised machine learning algorithms to detect fraudulent transactions in an anonymized dataset of over 200,000 records. To handle the <1 % fraud imbalance, the Synthetic Minority Oversampling Technique (SMOTE) was employed. Models including Logistic Regression, Random Forest, XGBoost, and ANN were evaluated using recall, precision, F1-score, and ROC-AUC metrics. Results show that ensemble methods, particularly XGBoost, delivered the highest detection accuracy and recall, proving most effective for combating financial fraud.
The work demonstrates that adaptive AI models can enhance security and reliability in modern financial systems.
This study applies supervised machine learning algorithms to detect fraudulent transactions in an anonymized dataset of over 200,000 records. To handle the <1 % fraud imbalance, the Synthetic Minority Oversampling Technique (SMOTE) was employed. Models including Logistic Regression, Random Forest, XGBoost, and ANN were evaluated using recall, precision, F1-score, and ROC-AUC metrics. Results show that ensemble methods, particularly XGBoost, delivered the highest detection accuracy and recall, proving most effective for combating financial fraud.
The work demonstrates that adaptive AI models can enhance security and reliability in modern financial systems.
Date
Jun 1, 2026 1:00 PM — 3:00 PM
Event
Location
Stanford University, 450 Serra Mall, Stanford, CA 94305
450 Serra Mall, Stanford, CA 94305
Josh Ibitoye will present his research on AI-based credit card fraud detection systems at the IEEE 2026 Conference in Stanford, California.
This session explores how machine learning and ensemble models like XGBoost can minimize undetected fraud and strengthen financial systems through adaptive analytics.
“Our goal is to help financial institutions deploy real-time, explainable, and intelligent fraud detection solutions.”
— Josh Ibitoye, IEEE 2026
